the fan in which there is little or no exposure of the wheel, while the latter will deliver a relatively greater volume against average restrictions or resistances met with in practice The following table, from tests on a multiblade fan, shows the difference
in efficiencies at various pressures or resistances at a given speed in revolutions per minute when 25 per cent of the diameter of wheel is exposed [col. (a)] and when a tangent to the wheel passing through the cutoff point is also parallel to the axis of discharge [col. (&)]. The same wheel and casing, except for the cutoff piece, were used in each case. Figures 5 and 6 show the feature being considered.
Efficiencies, per cent
FIG. 5. FlG- 6-
Spirals of fan casings.
pressure, Mechanical Volumetric Manometric
water (a) (W (a) (&) (a) .»)
• 0.6 40.0 40.0 396 383 79.4 76.8
1.0 47.2 49.5 335 340 94.6 94.6
1.1 48.2 51.0 305 322 96.2 98.5
1.2 48.5 51.5 262 301 96.0 100.7
1.3 47.0 51.5 189 271 94.7 102.4
1.4 41.0 50.0 116 215 96.0 103.7
When a fan operates against considerable restriction or resistance to the flow of air, there is a strong tendency in the air to leak back through the wheel at the point where it is about to leave the spiral to pass through the fan discharge. If the delivery is restricted, it leaks back through the wheel at the point of maximum pressure. The above table shows how the efficiencies drop off in the fan of Fig. 5 as compared with that of Fig. 6. The maximum pressure in the air is at the fan discharge, while from the bottom of the wheel to the cutoff point (that is, the exposed portion of the wheel) the air has only that pressure imparted to it by the blades in that section of the wheel. This produces leakage back into the wheel as soon as the pressure at the discharge builds up due to restriction. Experiment shows that cutoff points with very small clearances have a merit which exists only in fancy. The effect upon efficiency of a reasonably large clearance at this point is negligible; the amount of clearance may be 5 per cent of the wheel diameter in medium-sized fans. Most of the noise in a fan originates at the cutoff, the function and form of which make it like a reed that will give a musical note unless this is guarded against. A blunt, rigid construction as in Fig. 5, will keep the noise down, whereas small clearance and a sharp, knife-like edge will give a loud, shrill noise. In very-high-speed fans the cutoff is sometimes made of a material less subject to vibrations than metal, such as soft wood.
A common form of inlet still used on many casings has a diameter equal to (and in some cases less than) the inlet of the wheel, a casting or rolled angle being used to facilitate connection to a pipe, as in Figs. 7 and 8. With these forms the contraction of the entering stream of air constitutes a restricted area and is a source of loss, as it requires a higher velocity for a given volume of air. The shape of the inlet connection should be conformed to the natural path of the outer particles of air by means of a conical or bell-shaped inlet connection such as shown in Fig. 9. A connection in the form of a frustum of a cone is preferable from a manufacturing standpoint